Our work is aimed at elucidating the factors that are associated with the progression of cervical dysplasia to invasive cervical carcinoma. Our initial work focused on comparing the distribution of 27 HPV types in cervical dysplasia and carcinoma and showed that 1) there is strong correlation between HPV group and diagnostic category and 2) moderate dysplasias harbor different, so-called "oncogenic" HPV types than severe dysplasia and carcinoma. Continuing work is designed to characterize the biologic potential of lesions associated with "oncogenic" HPV types that are preferentially identified in dysplastic lesions but are not found in our cancers. Current screening methods refer these lesions for follow-up as "premalignant" when our data suggests that they may not be at high risk for progressing. The specific aims of this proposal are: 1) to accummulate and characterize cytologic and histologic samples of the full spectrum of cervical lesions according to HPV genotype; 2) perform a cross-sectional study that compares the full spectrum of cervical lesions stratified by diagnostic category and HPV type using key morphologic and genotypic analyses including clonality, ploidy, loss of heterozygosity for chromosome 3, physical state of HPV, analysis of mitotic figures and centromere characteristics as well as molecular cytogenetic testing (MLPA). The pattern of data for individual HPV types and diagnostic categories will be compared with those of the malignant prototype, that is, HPV 16-associated cancers. Millions of women are treated for dysplastic cervical lesions each year, most of which have little risk for progression to invasion. This encumbers enormous amounts of health care resources and generates considerable anxiety and convenience for the women. Targeting the key features that identify high risk lesions could increase the specificity of cancer screening while reducing the burden of treating women whose lesions will not progress.